KR101407449B1 - Retry Ratio based link adaptation method - Google Patents

Abstract

The present invention discloses a link adaptation method based on an RR (Retry Ratio) scheme that conforms to the IEEE 802.11 HHT / VHT standard, and which takes into account the collision problem caused by hidden nodes and an open loop structure that is simple to implement. The RR-based link adaptation method includes an initial value setting step, an initial MCS selection step, an RR calculation step, a comparison step of MCS, and an MCS index increase / decrease determination step.

Description

{Retry Ratio based link adaptation method}

The present invention relates to a link-less method, and more particularly to an apparatus and method for link-less method, in which a structure is simple by adopting an open-loop structure and hidden- (RR) based link adaptation method for setting an initial Modulation and Coding Selection (MCS) value considering a collision problem caused by a node.

The link adaptation algorithm is a method of increasing the transmission speed of the system, the capacity of the system and the frequency efficiency while maintaining the required quality of service (QoS). In this paper, we propose a hybrid adaptive modulation and coding (H-ARQ) scheme and a hybrid automatic repeat request (H-ARQ) scheme. Is a link adaptation method that determines the most suitable transmission scheme among MCS (Modulation and Coding Selection) levels.

Although the IEEE 802.11 (2007) standard does not specify LA (Link Adaptation), the IEEE 802.11n (HT) / ac (VHT) standard describes a close-loop method using MCS feedback in response to an MCS request have. This method is more complicated than the open-loop method in terms of actual implementation. It is meaningful in sounding Presentation Protocol Data Unit (PPDU) or sounding Neighbor Discovery Protocol (DPD) that supports all SD (spatial dimension) .

The open loop scheme includes algorithms (ARF / AARF, CARA-like) and SNR-based algorithms (CHARM, SGRAin, SGRA, ONOE) based on the received ACK frame before transmission at the transmitting terminal.

Among ARF / AARF algorithms based on ACK frame, ARF / AARF considers only receipt of ACK frames. Therefore, most of products adopting this method are adopting this method because of its simple implementation. However, since transmission failure due to frame collision is not considered, The transmission rate can not be guaranteed. CARA-like is a method to control transmission rate by looking at RTS / CTS frame and ACK frame considering hidden-problem and frame collision.

The SNR-based algorithm uses an RSSI (Receive Signal Strength Indicator) of the ACK frame received from the transmitter and assumes an UP / DOWN symmetric channel. However, the initial MCS (Modulation and Coding Scheme) Therefore, there is a disadvantage that the transmission loss is high.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and it is an object of the present invention to provide an RR (Retry Ratio) based on an open-loop structure conforming to the IEEE 802.11HT / VHT standard and a simple implementation, and a collision problem by a hidden node Link adaptation method.

According to an aspect of the present invention, there is provided an RL-based link adaptation method including an initial value setting step, an initial MCS selection step, an RR calculation step, a comparison step of MCS, and an MCS index increase / decrease determination step. The initial value setting step sets a reference value MD_th used for decreasing a modulation and coding selection index (MCS) and a reference value MU_th used for increasing an MCS index, and sets a signal to interference and noise ratio (SINR) ) Value (SINR) and theoretically calculates the SINR value (MCS_theo) at the point when the throughput is saturated. In the initial MCS selection step, an MCS index is selected from a STL (Saturation Throughput Look-up Table) using an SINR value (SINR) of the current frame and an SINR value (MCS_theo) at a theoretical saturation rate. The RR calculation step calculates an RR (Retry Ratio) using LRC (Long Retry Count) and SRC (Short Retry Count). The comparing step 140 of the MCS compares whether the MCS of the previous frame is larger than the MCS of the current frame. If the MCS of the previous frame is larger than the MCS of the current frame (Yes), the MCS index increase / decrease determination step 150 performs an increase decision step 150 of determining an MCS index to increase the MCS index, (No), the MCS index reduction decision step 160 of deciding the decrease of the MCS index is performed. In the MCS index increase / decrease determination step, the calculated RR value, a reference value (MD_th) used to decrease the MCS index, and a reference value (MU_th) used to increase the MCS index are used.

The RR-based link adaptation method according to the present invention has an advantage of being able to guarantee an optimized effective transmission rate since it is simple to implement and can adapt a fast link at the beginning of transmission and reception of a data frame, thereby minimizing loss of transmission amount.

Figure 1 shows an RR-based link adaptation method according to the present invention.
2 shows the relationship between the saturated transmission rate and the SINR value in the legacy mode.
FIG. 3 shows the relationship between the saturated transmission rate and the SINR value in the HT-mixed mode.
4 shows an example of STL.
FIG. 5 shows the time per execution step included in the MSDU.
FIG. 6 shows a result of computer simulation of the transmission rate according to the SNR in the AWGN environment by MCS.
FIG. 7 shows a computer simulation result on an ideal reference performance according to a channel environment.
8 compares an RR-based link adaptation method according to the present invention with another link adaptation method in an AWGN environment.
9 shows the difference between the reference performance value in the AWGN environment and the value obtained in the RR-based link adaptation method according to the present invention.
10 shows the difference between the reference performance value and the value obtained by the CINR-based technique in the AWGN environment.

In order to fully understand the present invention and the operational advantages of the present invention and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings, which are provided for explaining exemplary embodiments of the present invention, and the contents of the accompanying drawings.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like reference symbols in the drawings denote like elements.

Figure 1 shows an RR-based link adaptation method according to the present invention.

Referring to FIG. 1, an RR-based link adaptation method 100 according to the present invention includes an initial value setting step 110, an initial MCS selection step 120, an RR calculation step 130, an MCS index increase / And a transmission / reception end determination step 170.

The initial value setting step 110 sets a reference value MD_th used for decreasing a modulation and coding selection index (MCS) and a reference value MU_th used for increasing an MCS index, and outputs a signal to interference and noise ratio (SINR) Noise ratio (SINR) value and calculates the SINR value (MCS_theo) at the point when the throughput is theoretically saturated.

The initial MCS selection step 120 selects the MCS index from the STL using the SINR value (SINR) of the current frame and the SINR value (MCS_theo) at the theoretical saturation rate.

The RR calculation step 130 calculates an RR (Retry Ratio) using LRC (Long Retry Count) and SRC (Short Retry Count).

The comparison step 140 of the MCS compares whether the MCS (i-1) of the previous frame is larger than the MCS (i) of the current frame. If the variable i designating the order of the frame to be transmitted or received is 1, the previous frame NCS (0) may be set in the initial value setting step 110.

The MCS index increase / decrease determination step 150 or 160 determines whether the MCS index of the previous frame is greater than the MCS (i) of the current frame (Yes) ), And if the MCS (i-1) of the previous frame is smaller than the MCS (i) of the current frame (No), the MCS index reduction decision step 160 of determining the decrease of the MCS index is performed, Determine the increase or decrease. In the MCS index increase / decrease determination steps 150 and 160, the calculated RR value, the reference value MD_th used for decreasing the MCS index, and the reference value MU_th used for increasing the MCS index are used.

The MCS index increase determination step 150 performs a first comparison step 151, a second determination step 152, and an MCS index increase step 153.

The first comparing step 151 compares whether the calculated RR value is smaller than a reference value MU_th used to increase the MCS index. If the calculated RR value is smaller than the reference value MU_th used to increase the MCS index (Yes), the second determination step 152 determines whether the MCS index of the current frame is 7 (seven). If the MCS index of the current frame is not 7, the MCS index increase step 153 increases the value of the MCS index of the current frame. If the RR value calculated in the first comparing step 151 is larger than the reference value MU_th used for increasing the MCS index (No), the MCS index of the current frame is 7 in the second determining step 152, After increasing the value of the MCS index of the current frame in the index increasing step 153, the transmitting / receiving end determining step 170 is performed.

The decrease decision step 160 of the MCS index performs a first comparison step 161, a second decision step 162 and a decrease step 153. [

The first comparison step 161 compares whether the calculated RR value is smaller than a reference value MD_th used to decrease the MCS index. If the calculated RR value is smaller than the reference value MD_th used to decrease the MCS index (Yes), the second determination step 162 determines whether the MCS index of the current frame is 0 (zero). The MCS index reduction step 163 reduces the value of the MCS index of the current frame if the MCS index of the current frame is not zero. If the RR value calculated in the first comparing step 161 is larger than the reference value MD_th used for decreasing the MCS index (No), the MCS index of the current frame is 0 in the second determination step 162, After decreasing the value of the MCS index of the current frame in the index increasing step 163, the transmitting / receiving end determining step 170 is performed.

The transmission / reception termination determination step 170 determines whether the transmission process or the reception process is terminated in the MCS index increase / decrease determination step (150, 160). If it is determined that the transmission process or the reception process has not ended (No), the RR calculation step 130, the MCS comparison step 140 and the MCS index increase / decrease determination steps 150 and 160 are performed. If it is determined that the process has ended (Yes), the RR-based link adaptation method 100 ends.

If it is determined that the transmission process or the reception process has not ended (No), the state of the processed frame can be applied to the calculation by increasing the value of the variable (i) used for the calculation one by one.

2 shows the relationship between the saturated transmission rate and the SINR value in the legacy mode.

FIG. 3 shows the relationship between the saturated transmission rate and the SINR value in the HT-mixed mode.

In FIGS. 2 and 3, the legacy is according to the 802.11a standard, HT-mixed is according to the 802.11n standard, and 1024 bytes and 4096 bytes are the size of the PSDU (PHY Service Data Unit).

Referring to FIG. 2, the throughput of the MCS 2 is saturated from 10 Mbps. Theoretically, the SINR value (MCS_theo) at the time when the throughput is saturated is about 7 dB. Referring to FIG. 3, the MCS2 saturates at 117 Mbps. At this time, the SINR value (MCS_theo) at the time when the transmission rate is saturated theoretically becomes about 10 dB.

The STL described in the MCS selection step 120 is generated by dividing the SINR value at the time when the transmission rate is saturated by the size, mode and MCS of the PSDU (PHY Service Data Unit), and the mode is 802.11 a legacy according to the standard and HT-mixed according to the 802.11n standard.

4 shows an example of STL.

Referring to FIG. 4, it can be seen that the STL is a table generated by dividing the SINR value at the time of the transmission rate saturation by the size, mode and MCS of the PSDU (PHY Service Data Unit).

Hereinafter, effects of the present invention, which have been confirmed through computer simulation, will be described.

The theoretical maximum throughput (TMT) can be expressed by Equation (1).

Here, the MAC Service Data Unit (MSDU) is a MAC layer service data unit.

는 MSDU에 포함된 수행 단계별 시간을 나타낸다. here

Represents the time per execution step included in the MSDU.

FIG. 5 shows the time per execution step included in the MSDU.

Referring to FIG. 5, DIFS (Distributed Coordination Function Inter-Frame Spacing), BO (Back Off), RTS (Real Time Stamp), Short Inter-Frame Spacing (SIFS), CTS The time allocated to ACK (ACKnowledge) can be confirmed.

The length l of the PSDU (PHY Service Data Unit) and the transmission rate Sm according to the SINR can be expressed by Equation (2).

Here, PERm is described in a book (Digital Communications) jointly developed by John Proakis and Massoud Saleh, and is not a core idea of the present invention, and thus a detailed description thereof will be omitted.

In order to verify the effect of the present invention experimentally, it is assumed that the reference value (MD_th) used to reduce the MCS index in the simulation conditions is 20%, and the reference value (MU_th) used to increase the MCS index is 10%. The conditions of the simulation are as follows.

Fixed-point simulation;

PLCP: HT-mixed 20 MHz;

Rx mode: SISO;

Channel Estimation & smoothing: Smart smoothing (default);

Short GI: OFF;

MCS: 0 to 7;

PSDU: 1024 bytes;

Channel: AWGN;

TGNA / C / F model, Mobility: 60.0Km / h;

Environment speed: 10Km / h

FIG. 6 shows a result of computer simulation of the transmission rate according to the SNR in the AWGN environment by MCS.

It can be seen that the result shown in Fig. 6 is almost the same as the result shown in Fig.

FIG. 7 shows a computer simulation result on an ideal reference performance according to a channel environment.

Referring to FIG. 7, when the channel environment is TGn-A, there is a loss of about 500 Kbps to 1 Mbps compared to the case where the channel environment is AWGN. In the case where the channel environment is TGn-F, It can be seen that a transfer rate drop of 6 Mbps or more occurs at the maximum.

Where AWGN stands for Additive White Gaussian Noise and TGn stands for IEEE High Throughput Task Group 'n'.

8 compares an RR-based link adaptation method according to the present invention with another link adaptation method in an AWGN environment.

Referring to FIG. 8, the RR-based scheme according to the present invention has substantially the same value as the reference performance, but the ThPut_based scheme is significantly lower than the reference performance, and the CINR- However, it can be seen that the transmission rate is reduced by about 2 Mbps at the SNR of 4 dB.

9 shows the difference between the reference performance value in the AWGN environment and the value obtained in the RR-based link adaptation method according to the present invention.

10 shows the difference between the reference performance value and the value obtained by the CINR-based technique in the AWGN environment.

Referring to FIG. 9, in the RR-based link adaptation method according to the present invention, a maximum loss of about 250 Kbps occurs at an SNR of 22 dB, but the difference is within a range of 0.2 Mbps and an average transmission loss of 100 Kbps or less occurs.

Referring to FIG. 10, in the case of the CINR-based technique, a maximum loss of 2 Mbps occurs at an SNR of 4 dB and a transmission loss of about 0.4 Mbps occurs at a 22 dB portion.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments. It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope of the present invention.

110: initial value setting step 120: initial MCS selection step
130: RR calculation 140: Comparison step of MCS
150, 160: MCS index increase / decrease determination step
170: transmission / reception end judgment step

Claims (6)

A reference value MD_th used for decreasing a Modulation and Coding Selection (MCS) index and a reference value MU_th used for increasing an MCS index are set and a SINR (Signal to Interference and Noise Ratio) value (SINR) An initial value setting step 110 of calculating an SINR value (MCS_theo) at a point when the throughput is theoretically saturated;
An initial MCS selection step 120 for selecting an MCS index from a STL (Saturation Throughput Look-up Table) using an SINR value (SINR) of the current frame and an SINR value (MCS_theo) at the time when the transmission rate is theoretically saturated;
An RR calculation step (130) of calculating an RR (Retry Ratio) using LRC (Long Retry Count) and SRC (Short Retry Count);
Comparing (140) the MCS to compare whether the MCS of the previous frame is greater than the MCS of the current frame; And
If the MCS of the previous frame is larger than the MCS of the current frame (Yes), an increase determining step 150 of determining the increase of the MCS index is performed. If the MCS of the previous frame is smaller than the MCS of the current frame (No ) Comprises an MCS index increment / decrement decision step (150, 160) for performing a decrease decision step (160) of an MCS index determining a decrease of an MCS index,
In the MCS index increase / decrease determination steps 150 and 160, the calculated RR value, a reference value MD_th used to decrease the MCS index, and a reference value MU_th used to increase the MCS index are used,
In the STL,
The SINR value at the time when the transmission rate is saturated is generated by dividing the SINR value by the size, mode and MCS of the PSDU (PHY Service Data Unit)
Wherein the mode is legacy according to the 802.11a standard and HT-mixed according to the 802.11n standard. The method according to claim 1,
(170) for determining whether the transmission process or the reception process is completed when the MCS index increase / decrease determination step (150, 160) has been performed,
If the RR calculation step 130, the MCS comparison step 140 and the MCS index increase / decrease determination step 150, 160 are performed, if it is determined that the transmission process or the reception process has not ended, Based link adaptation method. 3. The method of claim 2, wherein the step of determining an increase in the MCS index (150)
A first comparing step (151) of comparing whether the calculated RR value is smaller than a reference value (MU_th) used to increase the MCS index;
A second determination step (152) of determining whether the calculated RR value is smaller than a reference value (MU_th) used for increasing the MCS index (Yes), and whether the MCS index of the current frame is 7 (seven); And
And an MCS index increasing step 153 for increasing the value of the MCS index of the current frame if the MCS index of the current frame is not 7,
If the calculated RR value is greater than the reference value MU_th used for increasing the MCS index in the first comparing step 151, the MCS index of the current frame is 7 (170) is performed after increasing the value of the MCS index of the current frame in the MCS index increasing step (153) and the transmitting / receiving end determining step (153). 4. The method of claim 2, wherein the step of determining (160)
A first comparing step (161) of comparing whether the calculated RR value is smaller than a reference value (MD_th) used for decreasing the MCS index;
A second determination step (162) of determining whether the calculated RR value is smaller than a reference value (MD_th) used for decreasing the MCS index (Yes), and whether the MCS index of the current frame is 0 (zero); And
And an MCS index decreasing step (163) of decreasing the value of the MCS index of the current frame if the MCS index of the current frame is not 0,
If the calculated RR value is greater than the reference value MD_th used for decreasing the MCS index (No) in the first comparing step 161, the MCS index of the current frame is set to 0 (170) is performed after decreasing the value of the MCS index of the current frame in the MCS index increasing step (163) and the MCS index increasing step (163). delete The method according to claim 1,
The SRC defaults to 7 (seven), and increases by 1 when the RTS transmission fails and when the transmission of data having a length less than or equal to the RTS threshold is failed,
Wherein the LRC is set to 4 (four) by default, and increases by 1 when data transmission with a length greater than an RTS threshold is failed.

Images